Skip to main navigation Skip to search Skip to main content

SEMG-based estimation of human arm force using regression model

  • Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The force estimation of human arm is one of the main problems while controlling biomechantronic system. It's an effective method to estimate the endpoint force of human arm through the surface electromyography (sEMG) produced by the contraction of the muscles. In this paper, sEMG-based force is estimated by both Bayesian Linear Regression (BLR) and Support Vector Regression (SVR) algorithms. Experimental results show that both models have high performances, with the average of RMS errors in BLR all below 2 N and in SVR mostly below 1 N. Meanwhile, the force estimated by the BLR model is highly linear correlated with the measured value. This paper also discusses the effect of force estimation when electrodes placed in different positions on arm. The performance is better while electrodes are placed on both forearm and upper arm.

Original languageEnglish
Title of host publication2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538637418
DOIs
StatePublished - 2 Jul 2017
Event2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017 - Macau, China
Duration: 5 Dec 20178 Dec 2017

Publication series

Name2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017
Volume2018-January

Conference

Conference2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017
Country/TerritoryChina
CityMacau
Period5/12/178/12/17

Keywords

  • Bayesian Linear Regression
  • Support Vector Regression
  • force estimation
  • surface electromyography

Fingerprint

Dive into the research topics of 'SEMG-based estimation of human arm force using regression model'. Together they form a unique fingerprint.

Cite this